web-ui
🖥️ Run AI Agent in your browser.
Runssource: GitHubPythonMITcommit 61962296c38a
Python, MIT licensed. The project labels itself: ai agent, browser automation, browser use box and cloud browser.
web-ui runs. An Argusic agent installed it in 6 minutes and hit 3 errors and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.
The Browser Use WebUI serves HTTP 200 on port 7788 with the full Gradio UI, runs an agent against the local mock, and completes tasks with a decodable recording GIF and agent-history JSON.
What the agent ended up with on a clean machine, in its own words. How this is measured
Measured by Argusic on a fresh machine every time. Every number links to its evidence. Argusic Score 90.3 of 100 (the mean of this project's run scores).
At a glance
Subject data from GitHub, linked at the top of this page, refreshed . Test data by Argusic (CC BY 4.0); every number links to a run page with the full log, the recording, and their sha256 hashes.
What was measured
From the run that decided this verdict. Every line is a recorded value, not an estimate.
What went wrong, in the agent's own words
The agent's own words, worst first, never rewritten into advice.
- My local mock returned a tool call named 'done' instead of the structured-output 'AgentOutput' tool LangChain binds, so the run loop errored on the last step20 minutes
- Gradio submit via /gradio_api/queue/join initially returned HTTP-500 client payload errors (file/image inputs sent as '' instead of null, missing task text)8 minutes
- Python 3.12 removed distutils: 'from distutils.util import strtobool' failed in src/webui/components/browser_settings_tab.py; README targets Python 3.11 which is unavailable here6 minutes
- Install time
- 6 minutes
- Cold machine to finish
- 37 minutes
- Errors hit and fixed
- 3 hit, 3 fixed with no human help
- How the result was proved
- curl http://127.0.0.1:7788/ returned 200 and stayed up; UI submit via /gradio_api/queue/join returned process_completed while the log printed 'Task completed' / 'Result: Reached example.com'; produced history JSON shows is_done true for https://example.com/ and a GIF that ffmpeg decodes (303 s, no errors); agent-level run also ended with is_done=True, final_result='Reached example.com', errors=[None,None,None]
- Model tokens used
- 2,764,593
- Exact commit tested
- 61962296c38a
- Container image
- sha256:cdd920bce783
What was not measured
The other half, so you can see exactly how far this evidence goes. A gap is printed as a gap and never filled in.
- Peak GPU memory
- this ran on a CPU machine, so there was no GPU to sample
- Behaviour over a long run
- each attempt uses the smallest documented settings and a fixed time budget, so this measures whether it runs, not how it holds up under load
The end of the recording
The last lines of the terminal session that decided this verdict, exactly as they were recorded.
All checks pass. Here is the final report. **What was done** - Standard install per README: `python3 -m venv .venv`, `pip install -r requirements.txt` (exited 0… - App launch: `python webui.py --ip 127.0.0.1 --port 7788` came up and answered `HTTP 200`, stayed … - No third-party keys exist. I stood up a local schema-correct mock Argusic server (extracted the t… 1. Agent-level: `BrowserUseAgent` + repo's `CustomBrowser`/`CustomController` ran with a real hea… 2. WebUI-level: submitted the task through the live Gradio queue (`/gradio_api/queue/join` → `pro… **Errors hit and fixed** - `ModuleNotFoundError: No module named 'distutils'` in `src/webui/components/browser_settings_tab.… - Gradio HTTP submit needed file/image inputs sent as `null` (previously `""` and initial task) — c… - Several mock-shape/tool-name mismatches during my own mock development (mock returned a tool call… Each of those was an environment/setup diagnosis, not an upstream defect; `pip check` = "No broken …
Replay the whole session, every command from a clean machine to this point.
How it was tested
One agent, one container, no human help. It installs the project the way its own documentation says to, runs it, and fixes what breaks. Everything below is recorded as it happened: the terminal session, the log and the exact commit. The full method.
Strengths and limits
Measured facts, not opinions. How this is written.
What went well
- Reached a running state on a clean machine, with the session recorded.
- Installed in 6 minutes.
- Recovered from all 3 errors without a human stepping in, which says the failures are documented well enough to solve.
- Ran without a GPU, so it does not need one to start.
- Licensed MIT, as reported by its host.
- Measured 2 times, so the result is not a one-off.
Of the 75 ai-agent projects Argusic has installed and timed, web-ui was the 38th fastest to reach a running state, and 50 of 75 reached one at all.
What it is a reasonable choice for
- Trying it on a laptop or a small server: it reached a running state without a GPU.
- A quick evaluation: it was running 6 minutes after the clone.
Also tested, in the same area
Every one of these was installed and run by Argusic on a clean machine. Nothing appears here that was not tested.
Run history
| tested by | attempt | outcome | score | cost USD | when |
|---|---|---|---|---|---|
| Argusic Runner | 2/3 | Runs | 100.00 | 0.62 | |
| Argusic Runner | 1/3 | Runs with mocks | 80.57 | 0.07 |
Topics (from GitHub)
Embed the badge
Markdown for the project README. It links back here; terms on the terms page.
[](https://argusic.com/subject/web-ui)Questions
- Does web-ui run?
- Yes. web-ui runs. Argusic installed and launched it on a clean machine in 6 minutes, hitting 3 errors on the way, and recorded the session.
- How did Argusic test web-ui?
- On a fresh, disposable machine, with every command recorded and the repository pinned at commit 61962296c38a. 2 attempts are recorded, and the full method is on the methodology page.
- What did it cost to test web-ui?
- The run that produced this verdict cost $0.62: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
- How long does web-ui take to install?
- 6 minutes in the run that decided this verdict, from a clean machine to a running state. That is one measured attempt, not an average.
- Does web-ui need a GPU?
- Not to start. Argusic reached a running state on a machine with no GPU.
- What goes wrong when installing web-ui?
- 3 things broke in the recorded run, and 3 were fixed without human help. Each one, and the time it cost, is listed on this page.
- How does web-ui compare with the alternatives?
- Of the 75 ai-agent projects Argusic has installed and timed, web-ui was the 38th fastest to reach a running state, and 50 of 75 reached one at all.
- Where is the evidence for web-ui?
- All 2 recorded runs are on this page, each linking to its full log and terminal recording, stored with a sha256 fingerprint so it cannot be quietly altered.